Sandhya 4

8/17/2019 Sandhya 4

1/7

Recent diarrhoeal illness and risk of lower

respiratory infections in children under theage of 5 years

Wolf-Peter Schmidt,1* Sandy Cairncross,1 Mauricio L Barreto,2 Thomas Clasen1 and Bernd Genser1,2

Accepted 12 February 2009

Background Children in low-income settings suffering from frequent diarrhoeaepisodes are also at a high risk of acute lower respiratory infections(ALRI). We explored whether this is due to common risk factors forboth conditions or whether diarrhoea can increase the risk of ALRI

directly.

Methods We used a dynamic time-to-event analysis of data from two largechild studies in low-income settings in Ghana and Brazil, with thecumulative diarrhoea prevalence over 2 weeks as the exposure and

severe ALRI as outcome. The analysis was adjusted for baseline riskof ALRI and diarrhoea, seasonality and age.

Results The child population from Ghana had a much higher risk of diar-rhoea, malnutrition and death than the children in Brazil. In thedata from Ghana, every additional day of diarrhoea within 2 weeksincreased the risk of ALRI by a factor of 1.08 (95% CI 1.00–1.15). Inaddition, we found a roughly linear relationship between thenumber of diarrhoea days over the last 28 days and the risk of ALRI. In the Ghana data, 26% of ALRI episodes may be due torecent exposure to diarrhoea. The Brazilian data gave no evidencefor an association between diarrhoea and ALRI.

Conclusion Diarrhoea may contribute substantially to the burden of ALRI inmalnourished child populations.

Keywords Diarrhoea, pneumonia, malnutrition, ALRI

IntroductionDiarrhoeal diseases and acute lower respiratory infec-tions (ALRI) are regarded as the leading proximatecauses of death among children in low-income

settings.1 It has been shown that children whosuffer from repeated or severe episodes of diarrhoea

are also at a higher risk of ALRI.2

However, it is notclear whether these conditions are causally related, or

whether their observed co-occurrence merely reflectsthe presence of common risk factors, for example a

weak immune system and malnutrition.3

Diarrhoeal disease is an established risk factor foracute weight loss, malnutrition and stunting.4–8

Likewise, malnutrition and failure to thrive are

known risk factors for ALRI.9,10 A causal link between

diarrhoeal disease and subsequent risk of ALRI is

therefore biologically plausible and would be of

* Corresponding author. Department of Infectious and Tropical

Diseases, London School of Hygiene and Tropical Medicine,

Keppel Street, WC1E 7HT London, UK.

E-mail: [email protected]

1 Department for Infectious and Tropical Diseases, London

School of Hygiene and Tropical Medicine, London, UK.2 Instituto de Saúde Coletiva, Federal University of Bahia,

Salvador, Brazil.

The online version of this article has been published under an open access model. Users are entitled to use, reproduce, disseminate, or display the open access

version of this article for non-commercial purposes provided that: the original authorship is properly and fully attributed; the Journal and Oxford University Press

are attributed as the original place of publication with the correct citation details given; if an article is subsequently reproduced or disseminated not in its entirety

but only in part or as a derivative work this must be clearly indicated. For commercial re-use, please contact [email protected]

Published by Oxford University Press on behalf of the International Epidemiological Association.

The Author 2009; all rights reserved. Advance Access publication 11 March 2009

International Journal of Epidemiology 2009;38:766–772

doi:10.1093/ije/dyp159

766

8/17/2019 Sandhya 4

2/7

considerable public health importance because effec-tive interventions targeting diarrhoeal diseases11

would then also contribute to a reduction of ALRI.Diarrhoea may also increase the risk of ALRI in theshort term by causing acute micronutrient loss, stresson the immune system, dehydration or immobiliza-tion, thereby creating a vulnerable period of increased

risk of infections. There is some evidence that ALRI isa common finding and cause of death of malnour-ished children hospitalized for severe diarrhoea.12–14

However, in hospital case series it can be difficult todistinguish between cause, effect and co-occurrencedue to common risk factors.

Longitudinal population-based studies are more sui-table to investigate the dependence of ALRI risk onrecent diarrhoea, but we were unable to identify astudy in the literature addressing this question. Theaim of this study was to explore whether diarrhoeaincreases the risk of ALRI in the short term using adynamic time-to-event analysis of data from two largechild studies in low-income settings.

Materials and methodsStudy populations

We used data from two large vitamin A trials innorthern Ghana15 and north-eastern Brazil,16 bothconducted in semi-arid low-income settings between1990 and 1991. The Ghana study took place in a poorrural setting with extended families living in com-pounds and practicing subsistence farming. Thestudy area was chosen due to a high prevalence of

vitamin A deficiency. A full census of all residents was carried out prior to the trial. All compounds inthe study area were included in the trial. All childrenaged 0–60 months living in these compounds wereeligible for inclusion (10% loss to follow up). TheBrazil trial was conducted in Serrinha, a town with30 000 inhabitants 170 km north-west of Bahia. At thebeginning of the study, children were identified by acensus of the poorest neighbourhoods in the area.Eligible children were between 6 and 48 months of age, and had no active xerophthalmia, recent measlesor recent vitamin A supplementation (10% loss tofollow up). In both trials children were followed upfor 1 year. Socio-economic status was poorer inGhana, where a large majority of households relied

on open wells, bore holes and rivers as watersource. In Brazil, over 95% of households had piped-in water supply and over 90% had electricity.

Diarrhoea exposure

In Ghana, weekly field worker visits measured theprevalence of diarrhoeal disease symptoms since thelast visit and respiratory signs; in Brazil field workers

visited up to three times a week. In the Ghana data-set, diarrhoea was defined based on mothers’ defini-tions and local disease perception. In the Brazil data

diarrhoea was defined as the occurrence of three ormore loose stools over 24 h.

It has been recognized that the adverse effects of diarrhoea such as malnutrition and death (and poten-tially ALRI) depend not so much on the number of diarrhoea episodes, but on the duration of these epi-sodes17 as well as the number of days with diarrhoea

over a certain period of time, also known as the long-itudinal prevalence.6–8 Restricting the analysis to(rare) prolonged episodes would have led to a lossof power. We therefore used as a measure of recentdiarrhoea experience the number of days with diar-rhoea in an individual during a fixed time windowlooking back from the date on which an episode of

ALRI occurred, subsequently named ‘index day’. Wedecided to apply a 2-week window for the primaryanalysis, since episodes of this duration have beenshown to be associated with adverse outcomes. In asensitivity analysis we also tested 4-week windows.Days that were preceded by

8/17/2019 Sandhya 4

3/7

variance estimation.21 Thus, by controlling for the over-all risk of diarrhoea and ALRI, which should representthe effect of common risk factors for the two conditions(e.g. socio-economic, environmental, water, hygiene)the model allowed focussing on the short-term tem-poral association.

We fitted a separate model for each stratum defined

by the ALRI rate, and then combined estimators of the stratum specific models to an overall estimate.20

In other words, each time an ALRI episode occurs inany individual (‘index day’), the model compares thediarrhoea exposure over the 14-day time window inthat individual with the diarrhoea exposure at thatcalendar time in other individuals with a similar over-all rate of ALRI and diarrhoea, adjusted for age.

We explored the impact of diarrhoea in the weeksprior to the index day by moving the time window tothe past starting on the day before the index day. Forexample, exposure to diarrhoea during the time

window closest to the index day was specified asthe sum of diarrhoea days from Day 15 to Day 1

prior to the index day. This window was moved to thepast in weekly intervals. For each time window at

weekly intervals we fitted a separate model.The population attributable fraction (PAF), i.e. the

proportion of ALRI episodes due to diarrhoea wascalculated using the formula PAF¼p0 (HR 1)/HR,

where p0 is the proportion of cases exposed and HR the adjusted hazard ratio.22

ResultsBasic epidemiological characteristics of the two studypopulations are shown in Table 1. The children in

Ghana had a higher risk of diarrhoea and longer epi-sodes. The mortality rate was more than 10 timeshigher in Ghana than in Brazil. Figure 1 shows thedistribution of the number of days with diarrhoeaover the past 14 days. In Brazil, 76% of all daysobserved were not preceded by any diarrhoea withinthe prior 14 days. In Ghana this figure was only 62%.Time periods with many diarrhoea days were muchmore common in Ghana. Nearly 4% of all days in theGhana study were preceded by at least 14 days of diarrhoea. As expected, the longitudinal prevalenceof diarrhoea and the incidence of ALRI in an indivi-dual child were correlated both in Ghana (r ¼0.15,

P < 0.001) and in Brazil (r ¼0.18, P < 0.001).

The results of the Cox regression analysis are shownin Figure 2. The graphs show the increase in the ALRIhazard rate with every additional day of diarrhoeaover 14 days during time windows at increasing dis-tance from the index day. In Ghana, every additionalday of diarrhoea during the 14 days preceding theindex day increased the hazard rate of ALRI bya factor of 1.08 (95% CI 1.00–1.15). There was noindication that the effect differed between differentcategories of ALRI risk. The effect also did notdepend on whether children without any ALRI

episode were excluded (hazard ratio ¼ 1.07, 95% CI0.99–1.14). We also found that additional adjust-ment for weight-for-age z -score (which can beregarded as a marker for nutritional status but also

for socio-economic and environmental health condi-tions) hardly changed this association (hazardratio¼1.07, 95% CI 1.0–1.15).

Moving the 14-day time window to the past by 1 week still revealed a trend towards an increasedhazard rate of ALRI (HR ¼ 1.05, 95% 0.98–1.13).Time windows further into the past did not showan obvious association between diarrhoea and ALRI.Due to decreasing power of the analysis for the time

windows further away from the index day, these esti-mates have very wide confidence intervals.

Table 1 Epidemiological characteristics of the two studypopulations

Ghana Brazil

Number of children 1877 1209

Person-years 1455 1104

Diarrhoea

Incidence rate/PYa 9.0 7.0

Longitudinal prevalence (%) 17 5

Mean duration of episodes (days) 6.1 2.7

ALRIb

Number of episodes 162 128

Incidence rate/PY 0.11 0.12

Malnutrition

Weight-for-age z -score< 2 (%) 30 13

Mortality

Deaths (n) 77 4

Mortality rate/1000 PY 52.9 3.6

aPY¼ person-years of observation.bDefined as increased breathing rate (more than 50 breaths perminute) plus presence of any danger sign (see text).

0

5

10

15

20

25

30

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

number of diarrhoea days in last 14 days

Brazil

Ghana

75

70

65

60

p e r c e n t a g e

Figure 1 Exposure to diarrhoea. Shown is the distributionof the number of diarrhoea days in the 14 days for allindividuals and days under observation

768 INTERNATIONAL JOURNAL OF EPIDEMIOLOGY

8/17/2019 Sandhya 4

4/7

In striking contrast to Ghana, the Brazilian datagave no evidence for an association between diarrhoeaover the last 14 days and ALRI for any of the time

windows (Figure 2b). The hazard ratios were veryclose to unity for the time windows close to theindex day. There was some indication of an associa-tion for the lag times further into the past, but—asthe wide confidence intervals suggest—this may easilybe due to chance.

To further examine the association between thenumber of days with diarrhoea and ALRI found inthe Ghana data, we treated the number of diarrhoeadays as a categorical variable and compared the ALRI

risk of each category with the risk in time windows without any diarrhoea. Figure 3a shows the associa-tion between the number of days with diarrhoea and

ALRI risk compared with 0 days of diarrhoea duringthe last 14 days. There appears to be a roughly linearassociation between the two conditions, at leastfrom53 days with diarrhoea over 14 days (asexpected, the Brazil data revealed no such trend—not shown). To explore whether this associationextends further into the past, we constructed a28-day window, also looking back from the first day

after the index day (Figure 3b). The graph suggests

that the association is also present for the longer time window.

Population attributable fraction

For Ghana, we estimated the population attributablefraction, i.e. the proportion of ALRI in the populationdue to short-term exposure to diarrhoea, by assumingthat diarrhoea affects ALRI risk predominantly within2 weeks. The study population was exposed to any diar-rhoea within 2 weeks during 38% of the person-time of observation (Figure 1). The proportion of ALRI episodesexposed (i.e. with any diarrhoea in the preceding2 weeks) was 68%. Any exposure to diarrhoea within2 weeks was associated with an adjusted hazard ratio of

HR ¼1.61 (95% CI: 0.89, 2.95). The estimated PAF is0.68 (1.611)/1.61¼26%, suggesting that 26% of

ALRI cases may be due to diarrhoea in the previous2 weeks.

DiscussionOur analysis suggests that diarrhoea contributed sub-stantially to the risk of ALRI within a few weeks of itsoccurrence in a child population in Ghana, but not in

1.08

1.05

1.02

0.980.96

1.021.02

1.10

0.90

H a z a r d R a t i o

Ghana

Index day

1.10

1.00

0.991.00

0.990.991.00

1.041.05

0.90

H a z a r d R a t i o

lag time (days)

lag time (days)

Index dayBrazil

1.10

1.00

(a)

(b)

–50 –43 –36 –29 –22 –15 –8 –1

–50 –43 –36 –29 –22 –15 –8 –1

Figure 2 The risk of ALRI depending on diarrhoea overa 14-day window at different lag times for (a) Ghana and(b) Brazil. The index day 0 is the day on which ALRIpresence or absence was ascertained. The first estimate onthe right (1.08 for Ghana) denotes the increase in the risk(hazard) of ALRI with every additional day with diarrhoeaduring the 14 days preceding the index day starting fromDay 1. The next estimate (1.05 for Ghana) denotes theincrease in the risk of ALRI with every additional diarrhoeaday during the 14 days from Day 8 to Day 21, and so on

Figure 3 The risk of ALRI in Ghana depending on thenumber of diarrhoea days during (a) the last 14 and (b) thelast 28 days prior to the index day (time window closest tothe index day). To avoid convergence problems, the numberof diarrhoea days during the 28-day window was collapsedinto groups of four

RECENT DIARRHOEAL ILLNESS 769

8/17/2019 Sandhya 4

5/7

a child population in Brazil. In Ghana, the relation-ship was roughly linear: every additional day of diar-rhoea over 2–4 weeks appeared to increase the riskof ALRI.

The striking socio-demographic and epidemiologicaldifferences between the two study populations maybe a plausible explanation for the contrasting find-

ings. The Ghana study included children under6 months who were excluded in Brazil. In Ghana,the mortality rate was more than 10 times higherand diarrhoea was more common, especially in longepisodes (Table 1). Malnutrition was more prevalentin the Ghana study site, as was clinical Vitamin A deficiency,15 which was mostly sub-clinical inBrazil.16 ALRI incidence was similar, but this islikely to be due to the three times more frequenthousehold visits in Brazil, which increased thechance of a child presenting with all respiratorysigns necessary to meet the case definition. Socio-eco-nomic status and water access were much morefavourable in the Brazilian households compared

with those included in the Ghana study.The higher risk profile in the Ghana study may sug-

gest that diarrhoea increases the risk of ALRI predo-minantly in malnourished children with a severediarrhoea burden, who are at high risk of death. Inline with our findings, hospital studies have shownthat ALRI is a common comorbidity and cause of death among malnourished children admitted withsevere diarrhoea.12–14 Hospital case series have manylimitations. The strength of our study is that longitu-dinal data allow a much better exploration and quan-tification of the temporal link, which is difficult toestablish in hospital studies. Our population-based

study also allowed calculation of the population attri-butable fraction, which in the case of Ghana suggestsa substantial contribution of diarrhoea to the burdenof ALRI.

Among the methodological limitations, recall biasand misclassification may be the most important.In Ghana, diarrhoea was defined based on local dis-ease concepts identified during extensive formativeresearch. The Brazil study applied the WHO standarddefinition. Both approaches have been used exten-sively in epidemiological studies, but do not fullyavoid misclassification and biased disease estimates.Recall bias could have increased the estimated hazardratio of ALRI for the diarrhoea time window immedi-

ately prior to the index day. Mothers caring for a child with ALRI may have been more inclined to reportdiarrhoea during the time since the last visit thanthose with a healthy child. The risk for recall biasmay be more pronounced in Ghana, where field

workers only visited once per week rather than twoto three times per week as in Brazil. Recall bias maytherefore explain some of the association betweendiarrhoea during the 7 days before a field worker

visit, but is unlikely to have affected the estimatesincluding diarrhoea days reported at the previous

visit or earlier where no ALRI was present. InGhana, we found a fairly linear dose–response rela-tionship between the number of diarrhoea days for upto 28 days of diarrhoea (Figure 3) which is wellbeyond 7 diarrhoea days and cannot be explainedby recall bias alone.

We used an outcome definition similar to the defi-

nitions recommended by WHO for the presumptivetreatment of ALRI. There are inherent methodologicallimitations in defining diarrhoea and ALRI in largefield studies.18,23 Thorough clinical examination andimaging by medical staff are often not possible in thefield, where presumptive treatment of signs likelyto be due to ALRI predominates. Due to the low spe-cificity and sensitivity of most surveillance methods,misclassification of disease status is unavoidable, even

when using (as in our analysis) stringent criteria forsevere ALRI resulting in comparatively low figures foroverall risk (Table 1).19 Non-differential misclassifica-tion of diarrhoea may have decreased the size of apotential association between the two conditions; dif-ferential misclassification may have biased the esti-mate in either direction.24 Misclassification of ALRImay have lowered the precision of the estimate.In some cases of severe diarrhoea, hypovolaemicor septic shock may cause metabolic acidosis withrapid breathing misclassified as ALRI, which couldhave inflated the effect sizes seen.3 However, wethink it is unlikely that this could account for a sub-stantial proportion of the estimate, since rapid breath-ing due to metabolic acidosis represents a relativelyrare, as well as short-lived and terminal sign of diar-rhoea, unlikely to be regularly picked up by weeklyfield worker visits. Also, metabolic acidosis is unlikely

to cause danger signs of ALRI, on which ALRI defini-tion was also dependent.By stratifying the analysis by the individual ALRI

risk, and adjusting for diarrhoea longitudinal preva-lence, we believe we have minimized the potential forconfounding by common underlying risk factors fordiarrhoea and ALRI. Residual confounding shouldhave affected the estimate at all time lags, not exclu-sively those close to the index day. To confirm thetemporal association between diarrhoea and subse-quent ALRI we reversed the time axis in the Ghanadata, by simply moving the time window forward intime from the index day. There was no evidence thatdiarrhoea in the 14 days after the index day was

associated with ALRI (HR ¼0.99, 95% CI 0.93–1.05), which lends supports to a causative role of diarrhoea.

On the whole, we believe that methodological lim-itations of the analysis are a possible, but not themost likely, reason for the pronounced differences inthe relationship between diarrhoea and ALRI identi-fied between the two child populations from Ghanaand Brazil. Differences in the demographic, nutri-tional and socio-economic profile are in our view amore plausible explanation. While we acknowledgethat methodological limitations potentially inflated


8/17/2019 Sandhya 4

6/7

the effect size, the population attributable fractionresulting from the Ghana data may on the otherhand underestimate the true effect of diarrhoea on

ALRI. Our analysis only allowed us to explore theshort-term association between diarrhoea and ALRI.The longer-term association between diarrhoea, mal-nutrition and subsequent risk of ALRI is difficult to

study and would require close follow up of a largenumber of children over several years and carefulconsideration of a wide range of potential confoun-ders, data that are usually not available in settings

where the problem is greatest. Despite high biologicalplausibility, it remains unknown whether the long-term effect of frequent and severe diarrhoea episodeson the nutritional status of children25 also increasesthe risk of ALRI. There is a need for further large-scale longitudinal studies to confirm our findingsand explore long-term effects.

Bearing in mind that ALRI and diarrhoea probablyrepresent the two most common proximate causes of child mortality worldwide, an association between the

two conditions as suggested by our analysis wouldhave considerable public health implications. Ouranalysis reinforces the idea that ALRI is largely theendpoint of a long chain of causal factors,26 includ-ing poverty, nutrition and environmental health

(water, sanitation, indoor air pollution). Accordingto our results, diarrhoea may be an important medi-ator in this causal chain in very poor settings.Global figures on causes of child mortality, whichoften place ALRI or pneumonia at the top of thelist,27 need to be seen in this light. While specific mea-sures to reduce ALRI incidence and case fatality such

as vaccination28,29 and case treatment30 proved effec-tive and worthwhile, this analysis underscores that theunderlying causes may be the ultimate problem.

FundingWellcome Trust, UK (WT082569AIA).

AcknowledgementsThe authors are grateful to Saul Morris and DavidRoss for providing data and comments; to LucySmith, Simon Cousens and Kim Mulholland for com-ments on the draft.

Conflict of interest: None declared.

References1 Black RE, Morris SS, Bryce J. Where and why are 10

million children dying every year? Lancet 2003;361:2226–34.

2 Fenn B, Morris SS, Black RE. Comorbidity in child-hood in northern Ghana: magnitude, associated factors,and impact on mortality. Int J Epidemiol 2005;34:368–75.

3 Mulholland K. Commentary: comorbidity as a factor inchild health and child survival in developing countries. Int J Epidemiol 2005;34:375–77.

4 Guerrant RL, Schorling JB, Mcauliffe JF, de Souza MA.Diarrhea as a cause and an effect of malnutrition: diar-rhea prevents catch-up growth and malnutritionincreases diarrhea frequency and duration. Am J Trop Med Hyg 1992;47:28–35.

5 Lutter CK, Mora JO, Habicht JP et al. Nutritional supple-mentation: effects on child stunting because of diarrhea. Am J Clin Nutr 1989;50:1–8.

6 Morris SS, Cousens SN, Kirkwood BR, Arthur P, Ross DA.Is prevalence of diarrhea a better predictor of subsequent

mortality and weight gain than diarrhea incidence? Am J Epidemiol 1996;144:582–88.

7 Rowland MG, Rowland SG, Cole TJ. Impact of infectionon the growth of children from 0 to 2 years in an urbanWest African community. Am J Clin Nutr 1988;47:134–38.

8 Torres AM, Peterson KE, de Souza AC, Orav EJ,Hughes M, Chen LC. Association of diarrhoea andupper respiratory infections with weight and heightgains in Bangladeshi children aged 5 to 11 years. Bull

World Health Organ 2000;78:1316–23.9 Savitha MR, Nandeeshwara SB, Pradeep Kumar MJ, ul-Haque F, Raju CK. Modifiable risk factors for acute lowerrespiratory tract infections. Indian J Pediatr 2007;74:477–82.

10 Caulfield LE, de Onis M, Blossner M, Black RE.Undernutrition as an underlying cause of child deathsassociated with diarrhea, pneumonia, malaria, andmeasles. Am J Clin Nutr 2004;80:193–98.

11 Laxminarayan R, Chow J, Shahid-Salles SA. Interventioncost-effectiveness: overview of main messages. In:Jamison DT, Breman JG, Measham AR (eds). Disease

KEY MESSAGES

Children in poor settings with frequent diarrhoea episodes are often found to be also at high risk of ALRI, but it is unclear whether this is due to common risk factors or whether diarrhoea can increasethe risk of ALRI directly.

In this analysis, we found that diarrhoea may increase the risk of ALRI over a vulnerable period of

2–4 weeks in malnourished child populations. The results suggest that prevention of diarrhoea maycontribute to a reduction in ALRI, the leading immediate cause of death in children.

RECENT DIARRHOEAL ILLNESS 771

8/17/2019 Sandhya 4

7/7

Control Priorities in Developing Countries. New York:Oxford University Press and The World Bank, 2006,pp. 35–58.

12 Mitra AK, Khan MR, Alam AN. Complications and out-come of disease in patients admitted to the intensive careunit of a diarrhoeal diseases hospital in Bangladesh. Trans R Soc Trop Med Hyg 1991;85:685–87.

13

Islam SS, Khan MU. Risk factors for diarrhoeal deaths: acase–control study at a diarrhoeal disease hospital inBangladesh. Int J Epidemiol 1986;15:116–21.

14 Sibal A, Patwari AK, Anand VK, Chhabra AK, Chandra D. Associated infections in persistent diarrhoea—anotherperspective. J Trop Pediatr 1996;42:64–67.

15 Ghana VAST Study Team. Vitamin A supplementationin northern Ghana: effects on clinic attendances,hospital admissions, and child mortality. Lancet1993;342:7–12.

16 Barreto ML, Santos LM, Assis AM et al. Effect of vitamin A supplementation on diarrhoea and acute lower-respiratory-tract infections in young children in Brazil. Lancet 1994;344:228–31.

17 Baqui AH, Black RE, Sack RB, Yunus MD, Siddique AK,Chowdhury HR. Epidemiological and clinical characteris-tics of acute and persistent diarrhoea in rural Bangladeshichildren. Acta Paediatr Suppl 1992;381:15–21.

18 Lanata CF, Rudan I, Boschi-Pinto C et al. Methodologicaland quality issues in epidemiological studies of acutelower respiratory infections in children in developingcountries. Int J Epidemiol 2004;33:1362–72.

19 Rudan I, Tomaskovic L, Boschi-Pinto C, Campbell H.Global estimate of the incidence of clinical pneumoniaamong children under five years of age. Bull World Health Organ 2004;82:895–903.

20 Kelly PJ, Lim LL. Survival analysis for recurrent eventdata: an application to childhood infectious diseases. Stat Med 2000;19:13–33.

21 Lin DY, Wei LJ. The robust inference for the Cox Propor-tional Hazard model. J Am Stat Assoc 1989;84:1074–78.

22 Bruzzi P, Green SB, Byar DP, Brinton LA, Schairer C.Estimating the population attributable risk for multiple

risk factors using case-control data. Am J Epidemiol1985;122:904–14.23 Morris SS, Cousens SN, Lanata CF, Kirkwood BR.

Diarrhoea—defining the episode. Int J Epidemiol 1994;23:617–23.

24 Copeland KT, Checkoway H, McMichael AJ,Holbrook RH. Bias due to misclassification in the estima-tion of relative risk. Am J Epidemiol 1977;105:488–95.

25 Checkley W, Buckley G, Gilman RH et al. Multi-countryanalysis of the effects of diarrhoea on childhood stunting. Int J Epidemiol 2009;37:816–30.

26 Mulholland K. Childhood pneumonia mortality—a per-manent global emergency. Lancet 2007;370:285–89.

27 Rudan I, Boschi-Pinto C, Biloglav Z, Mulholland K,Campbell H. Epidemiology and etiology of childhood

pneumonia. Bull World Health Organ 2008;86:408–16.28 Mulholland K, Smith PG, Broome CV, Gaye A, Whittle H,

Greenwood BM. A randomised trial of a Haemophilusinfluenzae type b conjugate vaccine in a developing coun-try for the prevention of pneumonia—ethical considera-tions. Int J Tuberc Lung Dis 1999;3:749–55.

29 Levine OS, O’Brien KL, Knoll M et al. Pneumococcal vacci-nation in developing countries. Lancet 2006;367:1880–82.

30 Marsh DR, Gilroy KE, Van de WR, Wansi E, Qazi S.Community case management of pneumonia: at a tippingpoint? Bull World Health Organ 2008;86:381–89.

Published by Oxford University Press on behalf of the International Epidemiological Association

The Author 2009; all rights reserved. Advance Access publication 3 April 2009

International Journal of Epidemiology 2009;38:772–774

doi:10.1093/ije/dyp170

Commentary: What is the role of co-morbidityin child mortality?Christa L Fischer Walker* and Robert E Black

Accepted 2 March 2009

Childhood infectious diseases remain the leadingcause of morbidity and mortality among children

under 5 years of age around the world. Morethan half of all child deaths can be attributed to thethree most lethal communicable diseases: pneumonia(including sepsis in neonates), diarrhoea andmalaria.1 Prevention and treatment of these infectiousdiseases should be the greatest public health prioritiesto improve the survival of children living in commu-nities characterized by extreme poverty, poor healthsystems and environmental contamination. Children

* Corresponding author. Department of International Health,

Johns Hopkins Bloomberg School of Public Health, 615

North Wolfe Street, Room E5535, Baltimore, MD 21205,

USA. E-mail: [email protected]

Department of International Health, Johns Hopkins Bloomberg

School of Public Health, Baltimore, MD, USA.


Documents

Sandhya 4